CA1266111A - Spacecraft camera image registration - Google Patents
Spacecraft camera image registrationInfo
- Publication number
- CA1266111A CA1266111A CA000525044A CA525044A CA1266111A CA 1266111 A CA1266111 A CA 1266111A CA 000525044 A CA000525044 A CA 000525044A CA 525044 A CA525044 A CA 525044A CA 1266111 A CA1266111 A CA 1266111A
- Authority
- CA
- Canada
- Prior art keywords
- spacecraft
- compensation signal
- camera
- mirror
- external
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Navigation (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Photometry And Measurement Of Optical Pulse Characteristics (AREA)
Abstract
Abstract SPACECRAFT CAMERA IMAGE REGISTRATION
A system for achieving spacecraft camera (1, 2) image registration comprises a portion external to the spacecraft and an image motion compensation system (IMCS) portion onboard the spacecraft. Within the IMCS, a computer (38) calculates an image registration compensation signal (60) which is sent to the scan control loops (84, 88, 94, 98) of the onboard cameras (1, 2). At the location external to the spacecraft, the long-term orbital and attitude perturbations on the spacecraft are modeled. Coefficients (K, A) from this model are periodically sent to the onboard computer (38) by means of a command unit (39). The coefficients (K, A) take into account observations of stars and landmarks made by the spacecraft cameras (1, 2) themselves, The computer (38) takes as inputs the updated coefficients (K, A) plus synchronization information indicating the mirror position (AZ, EL) of each of the spacecraft cameras (1, 2), operating mode, and starting and stopping status of the scan lines generated by these cameras (1, 2), and generates in response thereto the image registration compensation signal (60). The sources of periodic thermal errors on the spacecraft are discussed. The system is checked by calculating measurement residuals", the difference between the landmark and star locations predicted at the external location and the landmark and star locations as measured by the spacecraft cameras (1, 2).
A system for achieving spacecraft camera (1, 2) image registration comprises a portion external to the spacecraft and an image motion compensation system (IMCS) portion onboard the spacecraft. Within the IMCS, a computer (38) calculates an image registration compensation signal (60) which is sent to the scan control loops (84, 88, 94, 98) of the onboard cameras (1, 2). At the location external to the spacecraft, the long-term orbital and attitude perturbations on the spacecraft are modeled. Coefficients (K, A) from this model are periodically sent to the onboard computer (38) by means of a command unit (39). The coefficients (K, A) take into account observations of stars and landmarks made by the spacecraft cameras (1, 2) themselves, The computer (38) takes as inputs the updated coefficients (K, A) plus synchronization information indicating the mirror position (AZ, EL) of each of the spacecraft cameras (1, 2), operating mode, and starting and stopping status of the scan lines generated by these cameras (1, 2), and generates in response thereto the image registration compensation signal (60). The sources of periodic thermal errors on the spacecraft are discussed. The system is checked by calculating measurement residuals", the difference between the landmark and star locations predicted at the external location and the landmark and star locations as measured by the spacecraft cameras (1, 2).
Description
~x~
Desc~iDeion S?AC2CRAFT CA~E~A IMAG~ ?~EGIST IO~l Sta'ement Ot Gover_mental InterQst The invention described herein was made in tne performance of work under NASA contract no. MAS5-29500 and is subject ~o the provisions of 305 of tne ~a~ional .~eronautics and Space Act of 1953 (72 Stat.
~35; 12 ~SC 2-~7).
Technical_Field ~his invention pertains to the field of maintaining within a preselected limit the angular separation of corresponding pixels of repeated images of the same selected imaging area of a spacecraft camera.
3ackground Art U.S. patent 3,952,151 discloses a method and apparatus for s'abilizing an image produced by e.g., a camera on board a satellite, by sensing the instantaneous attitude displacement of the satellite and using these signals to adjust the image-generating beam at the ground station. ~he instant invention differs from the reference system in that: 1) It is a closed-loop system whereas the reference system is an open-loop system. 2) It does not require the gyroscopes needed by the reference system Gyroscopes ~5 are heavy, consume much power, are not very accurate, have a stability and drift problem, and normally require an onboard star tracker for calibration~ 3) It corrects for orbit and thermal variations as well as attitude control variations, whereas the reference system corrects for just attitude control variations.
5) It uses the camera 1, 2 itself to self-correct for errors whereas the reference system does not.
i6~
Secondary patent references are U.S. patents 3,~,3,777; 3,676,581; 3,716,669; 3,769,710; 3,359,460;
~,012,01S; and ~,300,159.
~ he following three items give a general description of portions of the invention:
1) D.W. Graul (one of the instant inventors), oral presentation accompanied by a posterboard displav before the 2nvironmental Research Institute of ~lichigan at its International Symposium on Remote Sensing of the 1~ Environment, October 21, 1985; 2) "New GOES to Sharpen Severe Weather Tracking~, Aviatlon Week and Space Technolo~y, December 23, 1985; and 3) A. Schwalb, ~Envirosat-2000 ~eport; GOES-Next Overview", National Oceanic and Atmospheric Administration, September, 1935 (pages 25, 26, 28, 32, 35, 36).
Disclosure of Invention One or more cameras (1, 2) onboard a spacecraft generates images of scenes external to the spacecraft, such as on the earth. Long term motion pertubations on the orbit and attitude of the spacecraft are determined as part of operations ground equipment (OGE). This information, in the form of coefficents (K, A), is periodically fed to an on-board computer (38), which, in response thereto, generates an image registration compensation signal (60). Signal (60) is sent to the servo control loops (84, 88, 94, 98) of the camera mirrors (33, 32).
Brief Descri~tlon of the Drawings These and other more detailed and specific objects and features of the present invention are more fully disclosed in the following specification, reference being had to the accompanying drawings, in which:
Figure 1 is an elevational view of a satellite which can advantageously employ the present invention;
Figure 2 is a sketch of the field of view of camera ~, 2 showing how the present invention achieves image motion compensation;
Figure 3 is a graph showing the declination of a geosynchronous satellite having a 0~ degree inclination;
Figure 4 is a graph showing N-S pixel shift within an 85-minute imaging or sounding interval as a function of scan azimuth, for a scan elevation of 0 degrees for the satellite of Fig. 3;
Figure 5 is a graph showing E-W pixel shift within an 85-minute imaging or sounding interval as a function of scan azimuth, for a scan elevation of 8 degrees for the satellite of Fig. 3;
Figure 6 is a graph showing N-S pixel shift within an 85-minute imaging or sounding interval as a function of scan azimuth, for a scan elevation of 8 degrees for the satellite of Fig. 3;
Figure 7 is a sketch showing how the present invention uses onboard step compensation to compensate a scan line; and Figure 8 is a functional block diagram of the image motion compensation system (IMCS) portion of the_ oresent invention. --Best Mode for Carrying Out the Invention Image registration is the process of limiting theeLror in the angular separation of corresponding pixels (with respect to each other) of repeated images of the same selected imaging area (frame) to within a specified preselected limit (for an example, see Table 1). The images are taken by one or more cameras 1, 2 onboard a spacecraft.
Although the present invention has utility on any type of spacecraft, it will be particularly illustrated with respect to the spacecraft shown in Fig. 1: one of the geosynchronous GOES IJKLM meteorological satellites sponsored by NOAA (National Oceanic and Atmospheric ~dministra~ion) and contracted for by ~T~SA (~lational ~eronautics and Space Administration). The items snown on ~ig. 1 include sola~ array 11, ~-ray sensor 12, magneto~etor 13, S-band transmit antenna 14, search and 5 rescue antenna 15, ~HF antenna 16, telemetry and command antenna 18, earth sensors 19, S-'oand receive antenna 20, solar sail 24, and two cameras: imager 1 and sounder 2. Imager 1 comprises cooler 17, aperture 23, and mirror 33. Sounder 2 comprises cooler 21, 10 aperture 22, and mirror 32.
The mirrors 33, 32 are each mounted on a two-axis gimbal which selectively positions the mirror 33, 32 with respect to orthogonal x and y axes at a very fast step-scan rate of many successive positions per second. The ostensibly common x axis can also be referred to as the roll, north/south, or ele'vation axis. The y axis for each mirror 33, 32 can also be referred to as the pitch, east/west, or azimuth axis.
Imager 1 provides multispectral radiometric imaging of the earth's surface, which can be useful, for example, in measuring transverse cloud velocity.
Imager 1 has five channels, four infrared and one visible; its two-axis gimbaled scanning mirror 33 sweeps an eight kilometer longitudinal swath along an east/west path on the earth, providing co-registered data of the viewed scene from all channels simultaneously. Position and size of the area scanned are controlled by command from scan logic 83, 87 (Fig.
8). The field of view of imager 1 is divided into a set of parallel scan lines each comprising many pixels~ The pixel size (on the earth) is as small as 1 km by 1 km for one of the channels. A scan frame (comprising many scan lines) is that subset of the total possible field of view that is commanded to be scanned. The scan frame is scanned in an amount of time known as the ~correlation time~, which is 22 minutes for a whole earth scan, less for an ~area scan"
- ~ -(~ortion of the ear~h). ~adio~etric calibration is ?rovided by ~eriodic mirror 33 slews to space and to ~n internal blackbody target.
Sounder 2 measures moisture content and temperature within the earth's atmosphere on a pixel-by-pixel basis. Sounder 2 comprises a 19 channel (18 infrared and l visible) discrete filter wheei radiometer; its two-a~is gimbaled scanning mirror 32 step-scans a 40 kilometer longitudinal swath across an east/west path-in 10 ~ilometer increments The nominal pixel size (onthe earth) is 10 km by 10 k~. ~ scan frame (comprising many scan lines) is scanned in an amount of time known as the correlation time. Passive radiation cooler ~1 controls the filter wheel assembly temperature. This allows operation at lower temperature or increased sensitivity. Radiometric calibration is provided b~v periodic mirror 32 slews to space and to an internal blackbody target.
Imager l and sounder 2 operate independently and ~0 simultaneously over a period o~ time known as an imaging or sounding linterval~. The interval is specified to be at least 85 minutes and can be as much as 12 hours. During an interval, several frames are scanned and several images made. At the end of the interval, the instruments 1, 2 are turned off and the spacecraft may enter the housekeeping mode, e.g., to fire thrusters for purposes of attitude control or momentum dumping.
The total pixel registration error is the sum of the N-S and E-W separation angles between centroids of corresponding pixels for successive images within an imaqing or sounding interval. Needless to say, the lower the pix~l registration error, the higher the quality of images emanating from instruments l, 2.
The present invention achieves the GOES IJXLM image registration accurac~ requirements (shown in Table l) with margin.
Table l.
GO~S IJXLM
Image Registration ~equirements in Microradians (3 sigma) Between two Within 35 min Consecutive Imaging/Sounding Imaging/Sounding Interval Intervals Image. 1 ~ast-~est ~2 336 North-So~th 42 336 Sounder 2 ~ast-West 66 33~
North~South 66 336 The present invention eliminates deterministic long-term motion effects on image registration in a closed-1002 real-time fashion using an onboard image motion compensation system (IMCS), which operates directly on the mirrors 33, 32, and a model of orbital ~0 and attitude motion located on the earth. The invention is transparent to the user. The mirror 33, 32 normal motion is slow enough for the compensation system's computer 38 to apply effective E-W (~zimuth) and N-S (elevation) corrections simultaneously.
The long-term motion effects compensated ~y the present invention include orbital inclination and orbital eccentricity (or~ital effects); and yaw error, structural thermal distortion, and earth sensor 19 thermal variation (attitude effects).
Figure 2 shows scan lines, with respect to a Mercator projection of the earth, traced by imager 1 or sounder 2 with and without pi~el shift due to orbit/attitude perturbations. The shift has been exaggerated in the Figure for purposes of 35 illustration. It is desired that the scan lines be horizontal In reality, mirrors 33, 32 Eollow arcuate paths to accomplish this, due to the earth's curvature Another consequence of the earth's ~6~
~urvature is that the sizes of the pixels o~ the earth varV, The scan line with pixel shift depicted in Fig. 2 is 3k~wed in both elevation and azimuth, although the 5 a~imuth s~ew can't be detected just by looking at ~ig.
Desc~iDeion S?AC2CRAFT CA~E~A IMAG~ ?~EGIST IO~l Sta'ement Ot Gover_mental InterQst The invention described herein was made in tne performance of work under NASA contract no. MAS5-29500 and is subject ~o the provisions of 305 of tne ~a~ional .~eronautics and Space Act of 1953 (72 Stat.
~35; 12 ~SC 2-~7).
Technical_Field ~his invention pertains to the field of maintaining within a preselected limit the angular separation of corresponding pixels of repeated images of the same selected imaging area of a spacecraft camera.
3ackground Art U.S. patent 3,952,151 discloses a method and apparatus for s'abilizing an image produced by e.g., a camera on board a satellite, by sensing the instantaneous attitude displacement of the satellite and using these signals to adjust the image-generating beam at the ground station. ~he instant invention differs from the reference system in that: 1) It is a closed-loop system whereas the reference system is an open-loop system. 2) It does not require the gyroscopes needed by the reference system Gyroscopes ~5 are heavy, consume much power, are not very accurate, have a stability and drift problem, and normally require an onboard star tracker for calibration~ 3) It corrects for orbit and thermal variations as well as attitude control variations, whereas the reference system corrects for just attitude control variations.
5) It uses the camera 1, 2 itself to self-correct for errors whereas the reference system does not.
i6~
Secondary patent references are U.S. patents 3,~,3,777; 3,676,581; 3,716,669; 3,769,710; 3,359,460;
~,012,01S; and ~,300,159.
~ he following three items give a general description of portions of the invention:
1) D.W. Graul (one of the instant inventors), oral presentation accompanied by a posterboard displav before the 2nvironmental Research Institute of ~lichigan at its International Symposium on Remote Sensing of the 1~ Environment, October 21, 1985; 2) "New GOES to Sharpen Severe Weather Tracking~, Aviatlon Week and Space Technolo~y, December 23, 1985; and 3) A. Schwalb, ~Envirosat-2000 ~eport; GOES-Next Overview", National Oceanic and Atmospheric Administration, September, 1935 (pages 25, 26, 28, 32, 35, 36).
Disclosure of Invention One or more cameras (1, 2) onboard a spacecraft generates images of scenes external to the spacecraft, such as on the earth. Long term motion pertubations on the orbit and attitude of the spacecraft are determined as part of operations ground equipment (OGE). This information, in the form of coefficents (K, A), is periodically fed to an on-board computer (38), which, in response thereto, generates an image registration compensation signal (60). Signal (60) is sent to the servo control loops (84, 88, 94, 98) of the camera mirrors (33, 32).
Brief Descri~tlon of the Drawings These and other more detailed and specific objects and features of the present invention are more fully disclosed in the following specification, reference being had to the accompanying drawings, in which:
Figure 1 is an elevational view of a satellite which can advantageously employ the present invention;
Figure 2 is a sketch of the field of view of camera ~, 2 showing how the present invention achieves image motion compensation;
Figure 3 is a graph showing the declination of a geosynchronous satellite having a 0~ degree inclination;
Figure 4 is a graph showing N-S pixel shift within an 85-minute imaging or sounding interval as a function of scan azimuth, for a scan elevation of 0 degrees for the satellite of Fig. 3;
Figure 5 is a graph showing E-W pixel shift within an 85-minute imaging or sounding interval as a function of scan azimuth, for a scan elevation of 8 degrees for the satellite of Fig. 3;
Figure 6 is a graph showing N-S pixel shift within an 85-minute imaging or sounding interval as a function of scan azimuth, for a scan elevation of 8 degrees for the satellite of Fig. 3;
Figure 7 is a sketch showing how the present invention uses onboard step compensation to compensate a scan line; and Figure 8 is a functional block diagram of the image motion compensation system (IMCS) portion of the_ oresent invention. --Best Mode for Carrying Out the Invention Image registration is the process of limiting theeLror in the angular separation of corresponding pixels (with respect to each other) of repeated images of the same selected imaging area (frame) to within a specified preselected limit (for an example, see Table 1). The images are taken by one or more cameras 1, 2 onboard a spacecraft.
Although the present invention has utility on any type of spacecraft, it will be particularly illustrated with respect to the spacecraft shown in Fig. 1: one of the geosynchronous GOES IJKLM meteorological satellites sponsored by NOAA (National Oceanic and Atmospheric ~dministra~ion) and contracted for by ~T~SA (~lational ~eronautics and Space Administration). The items snown on ~ig. 1 include sola~ array 11, ~-ray sensor 12, magneto~etor 13, S-band transmit antenna 14, search and 5 rescue antenna 15, ~HF antenna 16, telemetry and command antenna 18, earth sensors 19, S-'oand receive antenna 20, solar sail 24, and two cameras: imager 1 and sounder 2. Imager 1 comprises cooler 17, aperture 23, and mirror 33. Sounder 2 comprises cooler 21, 10 aperture 22, and mirror 32.
The mirrors 33, 32 are each mounted on a two-axis gimbal which selectively positions the mirror 33, 32 with respect to orthogonal x and y axes at a very fast step-scan rate of many successive positions per second. The ostensibly common x axis can also be referred to as the roll, north/south, or ele'vation axis. The y axis for each mirror 33, 32 can also be referred to as the pitch, east/west, or azimuth axis.
Imager 1 provides multispectral radiometric imaging of the earth's surface, which can be useful, for example, in measuring transverse cloud velocity.
Imager 1 has five channels, four infrared and one visible; its two-axis gimbaled scanning mirror 33 sweeps an eight kilometer longitudinal swath along an east/west path on the earth, providing co-registered data of the viewed scene from all channels simultaneously. Position and size of the area scanned are controlled by command from scan logic 83, 87 (Fig.
8). The field of view of imager 1 is divided into a set of parallel scan lines each comprising many pixels~ The pixel size (on the earth) is as small as 1 km by 1 km for one of the channels. A scan frame (comprising many scan lines) is that subset of the total possible field of view that is commanded to be scanned. The scan frame is scanned in an amount of time known as the ~correlation time~, which is 22 minutes for a whole earth scan, less for an ~area scan"
- ~ -(~ortion of the ear~h). ~adio~etric calibration is ?rovided by ~eriodic mirror 33 slews to space and to ~n internal blackbody target.
Sounder 2 measures moisture content and temperature within the earth's atmosphere on a pixel-by-pixel basis. Sounder 2 comprises a 19 channel (18 infrared and l visible) discrete filter wheei radiometer; its two-a~is gimbaled scanning mirror 32 step-scans a 40 kilometer longitudinal swath across an east/west path-in 10 ~ilometer increments The nominal pixel size (onthe earth) is 10 km by 10 k~. ~ scan frame (comprising many scan lines) is scanned in an amount of time known as the correlation time. Passive radiation cooler ~1 controls the filter wheel assembly temperature. This allows operation at lower temperature or increased sensitivity. Radiometric calibration is provided b~v periodic mirror 32 slews to space and to an internal blackbody target.
Imager l and sounder 2 operate independently and ~0 simultaneously over a period o~ time known as an imaging or sounding linterval~. The interval is specified to be at least 85 minutes and can be as much as 12 hours. During an interval, several frames are scanned and several images made. At the end of the interval, the instruments 1, 2 are turned off and the spacecraft may enter the housekeeping mode, e.g., to fire thrusters for purposes of attitude control or momentum dumping.
The total pixel registration error is the sum of the N-S and E-W separation angles between centroids of corresponding pixels for successive images within an imaqing or sounding interval. Needless to say, the lower the pix~l registration error, the higher the quality of images emanating from instruments l, 2.
The present invention achieves the GOES IJXLM image registration accurac~ requirements (shown in Table l) with margin.
Table l.
GO~S IJXLM
Image Registration ~equirements in Microradians (3 sigma) Between two Within 35 min Consecutive Imaging/Sounding Imaging/Sounding Interval Intervals Image. 1 ~ast-~est ~2 336 North-So~th 42 336 Sounder 2 ~ast-West 66 33~
North~South 66 336 The present invention eliminates deterministic long-term motion effects on image registration in a closed-1002 real-time fashion using an onboard image motion compensation system (IMCS), which operates directly on the mirrors 33, 32, and a model of orbital ~0 and attitude motion located on the earth. The invention is transparent to the user. The mirror 33, 32 normal motion is slow enough for the compensation system's computer 38 to apply effective E-W (~zimuth) and N-S (elevation) corrections simultaneously.
The long-term motion effects compensated ~y the present invention include orbital inclination and orbital eccentricity (or~ital effects); and yaw error, structural thermal distortion, and earth sensor 19 thermal variation (attitude effects).
Figure 2 shows scan lines, with respect to a Mercator projection of the earth, traced by imager 1 or sounder 2 with and without pi~el shift due to orbit/attitude perturbations. The shift has been exaggerated in the Figure for purposes of 35 illustration. It is desired that the scan lines be horizontal In reality, mirrors 33, 32 Eollow arcuate paths to accomplish this, due to the earth's curvature Another consequence of the earth's ~6~
~urvature is that the sizes of the pixels o~ the earth varV, The scan line with pixel shift depicted in Fig. 2 is 3k~wed in both elevation and azimuth, although the 5 a~imuth s~ew can't be detected just by looking at ~ig.
2 since the normal motion of the scan line is horizontal. ~he onboard compensation invoi~7es a step~ed sequence of incremental adjustments (~ AZ, 2L) that are applied to the azi~uth and elevation servo 10 control loops 88, 98, 84, 94 (~ig. ~3) while the mirrors 33, 32 are generating scan lines. The correction signal 60 is applied in a direction op~osite that of the undesired offset of the mirror 33, 32. In the example shown in Figure 2, the correction (~ AZ,Q EL) is applied in an east-south direction to compensate oppo~ing ~otion (west-north) of the mirror 33, 32.
This correction signal 60 is updated every 64 milliseconds along the scan line. When the compensated scan line is seen by the user, it appears very similar to a scan line with no pixel shift.
Further details of the compensation are illustrated in Fig. 7. Points 62 demarcate the beginning and end of each 64 msec compensation period. During each such period~ computer 38 calculates a ~AZ, ~ EL correction signal 60. Signal 60 is converted to analog form and low pass filtered by filters 81, 85, 91, 95, resulting in a smoothing of the compensation signal into the dashed form 63 depicted in Fig. 7. This smoothed signal 63 is then fed to the servo loops 84, 88, 94, 98, in four independent components 63A, B, C, D
corresponding to the two axes x, y of each of the two instruments 1, 2. The time constant of each low-pass filter 81, 8S, 91, 95 is selected such that the error in each smooth segment 63(n) is about 10% of the input step size ~EL(n), where n is a positive integer. n ranges from 1 through 14 to cover the entire scan line. In Fig. 7, n is shown ranging from 1 through 9.
~6~
?or imager 1, the maximum residual error after compensation is less than 2 mic~oradians. For s~under 2, the ~esidual error is even ~maller because of ~he slo~er stepping rate of mirror 32 compa~ed with mirror 5 33.
.~n orbital inclination perturbation ~ill no!~ be illustrated. Other long-term image shifts are similar. The spacecraft ground track (decllna~ion) due to a 0.1 inclination is shown in Figure 3. ~igures d 10 through 6 show t~e effect of the 0.1 inclination on pixel shift at two elevation angles (0, ~) at various times T (in hours) along the spacecraft ground track shown in ~igure 3. The east-west pixel shift for 0 elevation is not shown because it is zero. 8 is 15 considered an extreme elevation angle, because when the instrument 1, 2 is pointed at the earth, its maxi~rum elevation angle is 8.7 in either direction, for a total elevation range of 17.4 (when the instrument 1, 2 is pointing at stars, its maximum elevation range is ~0 21~).
Only a single 6~hour track is shown in Figs. 4-6 because the remaining 6-hour tracks are mirror images of the illustrated 6-hour track about the meridian or the equator. In Figs. 4-6, negative and positive indices after the symbol ~ indicate shifts at the beginning and end of the 85 minute imaging or sounding interval, respectively.
The constant E-W shift shown in Pigure 5 can be corrected relatively simply, by insertion of a fixed bias into signal 60 at the beginning of each scan line. More complicated shifts, such as the N-S shifts of ~igs. 4 and 6, are corrected as shown in Figure 7, by stepping along the scan line at the rate of once avery 64 ~sec (once per 1.25 for imager 1).
~igure 8 shows the hardware comprising the IMCS
portion of the invention. AOCE (attitude and orbit control electronics) processor 38 resides on board the g spacecraft; provides the azimuth and elevation corrections 60 to imager 1 and sounder 2 in four independent components 60 A, B, C, D corresponding to the two axes x, y of each of the two instruments 1, 2; and synchronizes the signal 60 magnitude with real-time information 99 indicating the beginning and end of each scan linel the direction of each scan line, the position (A2, EL) of each mirror 33, 32, and the operating mode of each camera 1, 2.
The azimuth corrections 60B, 60D and elevation corrections 60A, 60C are each a function of the mirror 33, 32 position (AZ, EL); a set of orbit/attitude compensation coefficients K, A; and synchronization data 99. Coefficients K, A are updated to processor 38 daily by command unit 39 associated with a ground portion of an image navigation system. A suitable image navigation system is more fully described in a commonly assigned U.S. PatQnt 4,688rO92, en~itled, "Satellite Camera Image Navigation".
The daily uplinked coefficients K, A are: seven orbital coefficients K (six orbital elements and one epoch time (beginning of a particular orbit/attitude variation)); and 25 attitude coefficients A (four for each of roll, pitch, and yaw of imager 1 and sounder 2, plus epoch time).
The six-element set of orbital coefficients A is used to provide orbit position predictions between coefficient updates. The attitude coefficients A are not simply instantaneous estimates of instrument 1, 2 roll, pitch, and yaw relative to the orbit, but are estimates of attitude model parameters, as will be explained in more detail below.
An additional 12 coefficients K, A are uplinked via ground command unit 39 during khe first imaging and sounding interval after each stationkeeping or eclipse~
using a quad~atic series in time or exponential .unction, respectively, as the attitude model.
Compensation signal 60 is converted to analog .orm by 10-bit D/A converters 41, 45, Sl, 55, then filtered 5 by low pass filters 81, 85, 91, 95. In the resulting analog signal 63, 21 microradians cor~esponds to one volt. This signal 63 has a maximum ~alue of +10 volts, which is sufficient for all required compensations during the 85-minute imaging or sounding interval. If-l0 signal 63 had a dynamic range of 6000 microradiansrather than 210 microradians, fixed gridding would be possible, i.e., pixels would have the same geographic coordinates for indefinitely long (and not just 85 minute) imaging and sounding intervals.
Preferably, a eedundant computer and D/A converters are provided to back up processor 38 and D/A converters 41, ~5, Sl, 55, respectively.
Mirror servos 44, 48, 5~, 58, which may comprise inductosyns, provide synchronization data buffer 61 20 with real-time information indicating the beginning and end of each scan line; the direction of each scan line;
the position (AZ, EL) of each mirror 33, 32; and the operating mode (normal, star sensing, area scan, space or blackbody calibration) of each camera 1, 2. This information is fed back from data buffer 61 to processor 38 in digital form 99 for purposes of synchronizing the compensation signal 60. For example, the AZ/EL portion of the feedback signal uses 32 bits for each mirror 33, 32.
The algorithm, residin~ in a ROM within computer 38, for calculating the compensation signal is as follows:
AZ = -A(Q)AZ~ Rs ~ 3(Q)~ ~s ~ EL(z-Ws)-y ~ EL = -A(Q)EL~ Rs ~ 3(Q)~a Ls + AZ(z-Ws)-x A(Q) = ~cosQ - C(Q)] 1 B(Q) = A(Q)C(Q) C(Q) = [cos ~ - 1 + (R~/RS) ] /
_os~ = cosAZcos~L
(~or star sighting, A(~ = 0, 3(Q) =
Re = earth's eadius = 6378.16 km Rs = geosynchronous radius = 42164.4 km 5 (~.~Z,~L) = pixel position compensation 60 components (azimuth, elevation) (A7~ ~L) = mirror 33, 32 position (azimuth, elevation) dRS = normalized radius deviation ~rom geosynchronous radius = e(sin?1-sinMI) 0 ~3 - longitude deviation from perigee crossing =
2e(sinM - sinMI) ~ s = satellite latitude = i(sinG - sinGI) Ws = satellite orbital rotation = i(cosG - cosGI) i = orbital inclination 15 G = argument of the latitude = M + w w = argument of perigee M = mean anomaly = n(t-to) + Mo MI = mean anomaly at middle of imaging/sounding interval = n(t-tI) + Mo ~0 e = eccentricity n = mean motion t = time to = time at epoch tI = time at middle of interval ~5 Mo ~ mean anomaly at epoch x, y, z = instrument l, 2 reference optical axes (roll, pitch, yaw) relative to middle of imaging/sounding interval.
Variables for the above algorithm are stored in a 30 RA.~ within processor 38.
The values of x, y, and z are each modeled as a trigonometric (harmonic) series in time. For example:
x = Alsinwt+A2coswt+A3sin2wt+A4cos2wt-(AlsinwtI+A2coswtI+A3sin2~tI+A4cos2wt~ ) These attitude models include terms up to the second harmonic in each of roll, pitch, and yaw for each instrument l, 2. wt represents the apparent daily ~6~
mo~ion of the sun about the spacecraft. Constant terms, if present, r~present fixed alignment biases.
~mplitudes of the harmonics (given b~ the coefficients ~1 through A4 in the above example) represent daily 5 variations due to solar radiation pressure effects on yaw, structural thermal distortions, and earth sensor 19 ~hermal drift. All these perturbations are periodic, and result from the sun's apparent daily motion about the s2acecraft. ~armonic series are used 10 to model the compensation algorithm because the effects being compensated are periodic. These 30urces of daily variation are not determined separately; rather, their collect~ve effects are characterized through the coefficients A in the trigonometric series.
~or the illustrated GOES IJKLM satellites, the orbit/attitude effects are measured by onboard star and landmark sensing, and by range information. The satellite attitude is nominally controlled by ~he use of magnetic torquers, earth sensors 19, and momentum ?O wheels.
The baseline IMCS digital logic has a linear range of +210 microradians with a resolution of 0.41 microeadians per least significant bit (LSB). The LSB
is defined by the resolution of the digital-to-analog 25 converters 41, 45, 51, 55.
The predicted repeatability error of the total IMCS
circuit over spacecraft life is less than 0.62 microradians. This 0.62 microradian repeatability error emanates from D/A circuits 41, 45, 51, 55; low 30 ~ass filters 81, 85, 91, 95; and buffer amplifiers 42, 46, 52, 56. The compensation algorithm is embedded in digital form within processor 38; therefore, this portion of the circuit is repeatable and has no drift effect due to aging. The nonlinearity of each ~/A
35 circuit 41, 45, 51, 55 is less than fO.05% ~aximum over the full input range. Since one LSB = 1/1024 (about 0.1%) of full scale, the +0.05~ is equivalent to +0.5 LSB a ' O . 71 microradians. This nonlinearity eEfect is a ~onthermal random pheno~enon which is not co~pensate~
by the presen~ invention.
Scan control loops 84, 88, 94, 98 are servo i error-correcting systems driven by commands emana.ing from mirror location command units 82, 86, 92, 96.
These co~mands tell the mirror 33, 32 the position (AZ, ~L) it should be in, based upon logic decisions made by scan cont~ol logic 83, 87, 93, 97. Feedback amplifiers 10 43, 47, 53, 57 comprise high stability resistive networks the variation oE gain of these amplifiers over their life is less than one LS8 (0~41 microradian).
The scan loop 84, 88, 94, ga response is deterministic, based on the following: knowledge of the 15 input signals to summing amplifiers 43, 47, 53, 57;
knowledge of the scan loop transfer function (response as a function of input); in-flight testing, conducted as part of the startup operation, to compare star and landmark locations wi~h compensation to their locations 20 without compensation; and continuous in-flight calibration of the IMCS as part of the overall scan loop calibration. The residual error in the scan loop 84, 88, 94, 98 is defined to be the response of the scan loop minus the inpu~ to the scan loop.
The total diurnal thermal error within the cameras 1, 2 is less tnan 0.741 microradian. This is a periodic attitude measurement effect and is thus compensated by the present invention.
The IMCS circuit was tested over the acceptance 30 test temperature range of -15C to + 40C for drift variation. The drift variation was shown to be less than one LSB. The tempe~ature coefficient of the D/A
converters 41, 45, 51, 55 is specified to be less than 15 ppm/C maximum. Since the diurnal thermal variation 35 on 3/A's 41, 45, 51, 55 is expected to be less than +5C daily, the drift variation of the D/A's due to te~pera~ure is 1~ X 10-6 X 5 = 0.0075%, equivalent to 0.07~ LSa = 0.031 microradian.
The imager l/sounder 2 buffer amplifiers ~2, 45, 52, 56, which receive the filtered analog com~ensation signals 63, are common mode rejection amplifiers each having a gain o~ l; each is followed by a high stability (4 picoradian/C) analog summing amplifier 43/ 47, 53, 57. The external temperature variation on these amplifier circuits during a daily cycle is lO anticipated to be less than +5C. Amplifier gains and stability are established to maintain deviation below 0-.71 microradian peak after compensation.
Although the above periodic externally-caused thermal errors are small, the present invention 15 compensates for their effects as part of the overall long-term compensation loop. This is made possible because observations of stars and eartn-based landmarks are made directly by the instruments 1, 2 themselves.
Star observations are typically made every 30 minutes, 20 and landmarks every 2 hours during daylight. The results of these observations, along with ranging measurements made by equipment located on the ground, are used to determine satellite orbit and instrument 1, 2 attitude, and, in turn, used to make periodic updates 25 to the coefficients K, A sent by ground command unit 39 to processor 38. Thus, satellite orbit/attitude effects are compensated by the compensation loop.
If the orbital and attitude perturbations were identical from one day to the next, a set of 30 coefficients K, A once determined would remain valid throughout the life of the spacecraft, and there would be no need ~or instruments 1, 2 to make periodic star and landmark observations. However, there are small day to day changes in these perturbations. These 35 changes result from: seasonal changes in yaw, thermal distortion, and earth sensor 1~ variations due to the sun's change in declination; and from secular drifts in mirro~ 33, 32 scan system performance due to such aging effec~s as variation in servo control loo~ voltage, servo gain, and component degradation.
3y continuall~ updating t'ne model coefficients X, A
5 through on-board star and landmark observ~tions, all of these effects are incorporated, r~sulting in a con~lnual calibration o~ the mirror 33, 32 compensation. This calibration does not disrupt any normal operations of the instruments 1, 2.
The operations ground equipment (OGE) comprises or~it and attitude estimation software, which computes predicted landmark and star locations and ~measurement residualsl. Measurement residuals are the difference between the landmark and star locations predicted by 15 the OGE, on the one hand, and the landmark and star locations as measured by cameras 1, 2, on the other hand. The calculation of the measurement residuals uses the same model and parameters that processor 38 uses in generating the image motion compensation signal 20 60, thus insuring that the residuals properly reflect the true day-to-day changes in the orbit and attitude perturbations. The generation of measurement residuals requires that compensation signal 60 be applied during star and landmark sightings as well as during earth 25 scans. During a star or landmark sighting, processor 38 applies the compensation voltage 60 appropriate to the time t of the star or landmark sighting and its position (AZ, EL) in the instrument aperture 23, 22.
In order to serve as a meaningful quality check of the entire range of the compensation signal 60, the stars and landmarks should be well distributed over the entire instrument 1, 2 field of view.
The OGE provides a continual quality monitoring of the compensation system. With a properly operating system, because of the frequency of star sightings and attitude model updates, the measurement residuals will normally be small. Any continued increase in these residual above a controllable preselected threshold causes an alert message to be genPrated.
The OGE provides an additional continuous quality check on the invention as follows. As part of the data stream continuously telemetered to the ground is sent the compensation signal 60; the AZ, ~L of each mirror 33, 32; and the error (feedback) signal leaving each servo 44, 48, 54, 58. the OGE computer has stored therein a duplicate of the compensation algorithm embedded within processor 38. This duplicate algorithm takes the telemetered information and calculates AZ and ~ EL, then compares these calculated values of ~ AZ and ~ EL
with the telemetered values of ~ AZ and ~ EL. They should coincide if the system is operating properly.
The present invention addresses the problem of long-term errors that impact image registration. Short-term stability errors are also present, and arise from two sources: spacecraft platform stability errors, and ~0 scan repeatability errors. The spacecraft platform stability errors have three sources: motion of the spacecraft in pitch and roll due to Qarth sensor 19 noise (the primary error), errors from mirror 33, 32 motive interaction, and the effects of solar array 11 drive operation.
Scan repeatability errors are due to fixed pattern noise of the inductosyn servos 44, 48, 54, 58; one-cycle errors; sine/cosine unbalances; second harmonic erxors;
bearing noises; bearing friction; wire drag; and servo transient errors.
The mirror 33, 32 interaction errors and solar array ll drive effect errors can be compensated for by the invention known as "Pointing Compensation System for Spacecraft Instruments: described in U.S. Patent No~
4,687,161, commonly assigned with the instant invention.
Spacecraft motion compensation logic 25 of said patent ~2~
application can be implemented as part of processor 33 descriDed herein.
~ he above description is included to illustrate the operation of the 2referred embodiments and is not meant S to limit the scope of the invention. The scope of the invention is to be limited only by the Eollowing claims. From the above discussion, many variations will be apparent to one skilled in the art that would yet be encompassed by the spirit and scope of the 10 invention.
What is claimed is:
This correction signal 60 is updated every 64 milliseconds along the scan line. When the compensated scan line is seen by the user, it appears very similar to a scan line with no pixel shift.
Further details of the compensation are illustrated in Fig. 7. Points 62 demarcate the beginning and end of each 64 msec compensation period. During each such period~ computer 38 calculates a ~AZ, ~ EL correction signal 60. Signal 60 is converted to analog form and low pass filtered by filters 81, 85, 91, 95, resulting in a smoothing of the compensation signal into the dashed form 63 depicted in Fig. 7. This smoothed signal 63 is then fed to the servo loops 84, 88, 94, 98, in four independent components 63A, B, C, D
corresponding to the two axes x, y of each of the two instruments 1, 2. The time constant of each low-pass filter 81, 8S, 91, 95 is selected such that the error in each smooth segment 63(n) is about 10% of the input step size ~EL(n), where n is a positive integer. n ranges from 1 through 14 to cover the entire scan line. In Fig. 7, n is shown ranging from 1 through 9.
~6~
?or imager 1, the maximum residual error after compensation is less than 2 mic~oradians. For s~under 2, the ~esidual error is even ~maller because of ~he slo~er stepping rate of mirror 32 compa~ed with mirror 5 33.
.~n orbital inclination perturbation ~ill no!~ be illustrated. Other long-term image shifts are similar. The spacecraft ground track (decllna~ion) due to a 0.1 inclination is shown in Figure 3. ~igures d 10 through 6 show t~e effect of the 0.1 inclination on pixel shift at two elevation angles (0, ~) at various times T (in hours) along the spacecraft ground track shown in ~igure 3. The east-west pixel shift for 0 elevation is not shown because it is zero. 8 is 15 considered an extreme elevation angle, because when the instrument 1, 2 is pointed at the earth, its maxi~rum elevation angle is 8.7 in either direction, for a total elevation range of 17.4 (when the instrument 1, 2 is pointing at stars, its maximum elevation range is ~0 21~).
Only a single 6~hour track is shown in Figs. 4-6 because the remaining 6-hour tracks are mirror images of the illustrated 6-hour track about the meridian or the equator. In Figs. 4-6, negative and positive indices after the symbol ~ indicate shifts at the beginning and end of the 85 minute imaging or sounding interval, respectively.
The constant E-W shift shown in Pigure 5 can be corrected relatively simply, by insertion of a fixed bias into signal 60 at the beginning of each scan line. More complicated shifts, such as the N-S shifts of ~igs. 4 and 6, are corrected as shown in Figure 7, by stepping along the scan line at the rate of once avery 64 ~sec (once per 1.25 for imager 1).
~igure 8 shows the hardware comprising the IMCS
portion of the invention. AOCE (attitude and orbit control electronics) processor 38 resides on board the g spacecraft; provides the azimuth and elevation corrections 60 to imager 1 and sounder 2 in four independent components 60 A, B, C, D corresponding to the two axes x, y of each of the two instruments 1, 2; and synchronizes the signal 60 magnitude with real-time information 99 indicating the beginning and end of each scan linel the direction of each scan line, the position (A2, EL) of each mirror 33, 32, and the operating mode of each camera 1, 2.
The azimuth corrections 60B, 60D and elevation corrections 60A, 60C are each a function of the mirror 33, 32 position (AZ, EL); a set of orbit/attitude compensation coefficients K, A; and synchronization data 99. Coefficients K, A are updated to processor 38 daily by command unit 39 associated with a ground portion of an image navigation system. A suitable image navigation system is more fully described in a commonly assigned U.S. PatQnt 4,688rO92, en~itled, "Satellite Camera Image Navigation".
The daily uplinked coefficients K, A are: seven orbital coefficients K (six orbital elements and one epoch time (beginning of a particular orbit/attitude variation)); and 25 attitude coefficients A (four for each of roll, pitch, and yaw of imager 1 and sounder 2, plus epoch time).
The six-element set of orbital coefficients A is used to provide orbit position predictions between coefficient updates. The attitude coefficients A are not simply instantaneous estimates of instrument 1, 2 roll, pitch, and yaw relative to the orbit, but are estimates of attitude model parameters, as will be explained in more detail below.
An additional 12 coefficients K, A are uplinked via ground command unit 39 during khe first imaging and sounding interval after each stationkeeping or eclipse~
using a quad~atic series in time or exponential .unction, respectively, as the attitude model.
Compensation signal 60 is converted to analog .orm by 10-bit D/A converters 41, 45, Sl, 55, then filtered 5 by low pass filters 81, 85, 91, 95. In the resulting analog signal 63, 21 microradians cor~esponds to one volt. This signal 63 has a maximum ~alue of +10 volts, which is sufficient for all required compensations during the 85-minute imaging or sounding interval. If-l0 signal 63 had a dynamic range of 6000 microradiansrather than 210 microradians, fixed gridding would be possible, i.e., pixels would have the same geographic coordinates for indefinitely long (and not just 85 minute) imaging and sounding intervals.
Preferably, a eedundant computer and D/A converters are provided to back up processor 38 and D/A converters 41, ~5, Sl, 55, respectively.
Mirror servos 44, 48, 5~, 58, which may comprise inductosyns, provide synchronization data buffer 61 20 with real-time information indicating the beginning and end of each scan line; the direction of each scan line;
the position (AZ, EL) of each mirror 33, 32; and the operating mode (normal, star sensing, area scan, space or blackbody calibration) of each camera 1, 2. This information is fed back from data buffer 61 to processor 38 in digital form 99 for purposes of synchronizing the compensation signal 60. For example, the AZ/EL portion of the feedback signal uses 32 bits for each mirror 33, 32.
The algorithm, residin~ in a ROM within computer 38, for calculating the compensation signal is as follows:
AZ = -A(Q)AZ~ Rs ~ 3(Q)~ ~s ~ EL(z-Ws)-y ~ EL = -A(Q)EL~ Rs ~ 3(Q)~a Ls + AZ(z-Ws)-x A(Q) = ~cosQ - C(Q)] 1 B(Q) = A(Q)C(Q) C(Q) = [cos ~ - 1 + (R~/RS) ] /
_os~ = cosAZcos~L
(~or star sighting, A(~ = 0, 3(Q) =
Re = earth's eadius = 6378.16 km Rs = geosynchronous radius = 42164.4 km 5 (~.~Z,~L) = pixel position compensation 60 components (azimuth, elevation) (A7~ ~L) = mirror 33, 32 position (azimuth, elevation) dRS = normalized radius deviation ~rom geosynchronous radius = e(sin?1-sinMI) 0 ~3 - longitude deviation from perigee crossing =
2e(sinM - sinMI) ~ s = satellite latitude = i(sinG - sinGI) Ws = satellite orbital rotation = i(cosG - cosGI) i = orbital inclination 15 G = argument of the latitude = M + w w = argument of perigee M = mean anomaly = n(t-to) + Mo MI = mean anomaly at middle of imaging/sounding interval = n(t-tI) + Mo ~0 e = eccentricity n = mean motion t = time to = time at epoch tI = time at middle of interval ~5 Mo ~ mean anomaly at epoch x, y, z = instrument l, 2 reference optical axes (roll, pitch, yaw) relative to middle of imaging/sounding interval.
Variables for the above algorithm are stored in a 30 RA.~ within processor 38.
The values of x, y, and z are each modeled as a trigonometric (harmonic) series in time. For example:
x = Alsinwt+A2coswt+A3sin2wt+A4cos2wt-(AlsinwtI+A2coswtI+A3sin2~tI+A4cos2wt~ ) These attitude models include terms up to the second harmonic in each of roll, pitch, and yaw for each instrument l, 2. wt represents the apparent daily ~6~
mo~ion of the sun about the spacecraft. Constant terms, if present, r~present fixed alignment biases.
~mplitudes of the harmonics (given b~ the coefficients ~1 through A4 in the above example) represent daily 5 variations due to solar radiation pressure effects on yaw, structural thermal distortions, and earth sensor 19 ~hermal drift. All these perturbations are periodic, and result from the sun's apparent daily motion about the s2acecraft. ~armonic series are used 10 to model the compensation algorithm because the effects being compensated are periodic. These 30urces of daily variation are not determined separately; rather, their collect~ve effects are characterized through the coefficients A in the trigonometric series.
~or the illustrated GOES IJKLM satellites, the orbit/attitude effects are measured by onboard star and landmark sensing, and by range information. The satellite attitude is nominally controlled by ~he use of magnetic torquers, earth sensors 19, and momentum ?O wheels.
The baseline IMCS digital logic has a linear range of +210 microradians with a resolution of 0.41 microeadians per least significant bit (LSB). The LSB
is defined by the resolution of the digital-to-analog 25 converters 41, 45, 51, 55.
The predicted repeatability error of the total IMCS
circuit over spacecraft life is less than 0.62 microradians. This 0.62 microradian repeatability error emanates from D/A circuits 41, 45, 51, 55; low 30 ~ass filters 81, 85, 91, 95; and buffer amplifiers 42, 46, 52, 56. The compensation algorithm is embedded in digital form within processor 38; therefore, this portion of the circuit is repeatable and has no drift effect due to aging. The nonlinearity of each ~/A
35 circuit 41, 45, 51, 55 is less than fO.05% ~aximum over the full input range. Since one LSB = 1/1024 (about 0.1%) of full scale, the +0.05~ is equivalent to +0.5 LSB a ' O . 71 microradians. This nonlinearity eEfect is a ~onthermal random pheno~enon which is not co~pensate~
by the presen~ invention.
Scan control loops 84, 88, 94, 98 are servo i error-correcting systems driven by commands emana.ing from mirror location command units 82, 86, 92, 96.
These co~mands tell the mirror 33, 32 the position (AZ, ~L) it should be in, based upon logic decisions made by scan cont~ol logic 83, 87, 93, 97. Feedback amplifiers 10 43, 47, 53, 57 comprise high stability resistive networks the variation oE gain of these amplifiers over their life is less than one LS8 (0~41 microradian).
The scan loop 84, 88, 94, ga response is deterministic, based on the following: knowledge of the 15 input signals to summing amplifiers 43, 47, 53, 57;
knowledge of the scan loop transfer function (response as a function of input); in-flight testing, conducted as part of the startup operation, to compare star and landmark locations wi~h compensation to their locations 20 without compensation; and continuous in-flight calibration of the IMCS as part of the overall scan loop calibration. The residual error in the scan loop 84, 88, 94, 98 is defined to be the response of the scan loop minus the inpu~ to the scan loop.
The total diurnal thermal error within the cameras 1, 2 is less tnan 0.741 microradian. This is a periodic attitude measurement effect and is thus compensated by the present invention.
The IMCS circuit was tested over the acceptance 30 test temperature range of -15C to + 40C for drift variation. The drift variation was shown to be less than one LSB. The tempe~ature coefficient of the D/A
converters 41, 45, 51, 55 is specified to be less than 15 ppm/C maximum. Since the diurnal thermal variation 35 on 3/A's 41, 45, 51, 55 is expected to be less than +5C daily, the drift variation of the D/A's due to te~pera~ure is 1~ X 10-6 X 5 = 0.0075%, equivalent to 0.07~ LSa = 0.031 microradian.
The imager l/sounder 2 buffer amplifiers ~2, 45, 52, 56, which receive the filtered analog com~ensation signals 63, are common mode rejection amplifiers each having a gain o~ l; each is followed by a high stability (4 picoradian/C) analog summing amplifier 43/ 47, 53, 57. The external temperature variation on these amplifier circuits during a daily cycle is lO anticipated to be less than +5C. Amplifier gains and stability are established to maintain deviation below 0-.71 microradian peak after compensation.
Although the above periodic externally-caused thermal errors are small, the present invention 15 compensates for their effects as part of the overall long-term compensation loop. This is made possible because observations of stars and eartn-based landmarks are made directly by the instruments 1, 2 themselves.
Star observations are typically made every 30 minutes, 20 and landmarks every 2 hours during daylight. The results of these observations, along with ranging measurements made by equipment located on the ground, are used to determine satellite orbit and instrument 1, 2 attitude, and, in turn, used to make periodic updates 25 to the coefficients K, A sent by ground command unit 39 to processor 38. Thus, satellite orbit/attitude effects are compensated by the compensation loop.
If the orbital and attitude perturbations were identical from one day to the next, a set of 30 coefficients K, A once determined would remain valid throughout the life of the spacecraft, and there would be no need ~or instruments 1, 2 to make periodic star and landmark observations. However, there are small day to day changes in these perturbations. These 35 changes result from: seasonal changes in yaw, thermal distortion, and earth sensor 1~ variations due to the sun's change in declination; and from secular drifts in mirro~ 33, 32 scan system performance due to such aging effec~s as variation in servo control loo~ voltage, servo gain, and component degradation.
3y continuall~ updating t'ne model coefficients X, A
5 through on-board star and landmark observ~tions, all of these effects are incorporated, r~sulting in a con~lnual calibration o~ the mirror 33, 32 compensation. This calibration does not disrupt any normal operations of the instruments 1, 2.
The operations ground equipment (OGE) comprises or~it and attitude estimation software, which computes predicted landmark and star locations and ~measurement residualsl. Measurement residuals are the difference between the landmark and star locations predicted by 15 the OGE, on the one hand, and the landmark and star locations as measured by cameras 1, 2, on the other hand. The calculation of the measurement residuals uses the same model and parameters that processor 38 uses in generating the image motion compensation signal 20 60, thus insuring that the residuals properly reflect the true day-to-day changes in the orbit and attitude perturbations. The generation of measurement residuals requires that compensation signal 60 be applied during star and landmark sightings as well as during earth 25 scans. During a star or landmark sighting, processor 38 applies the compensation voltage 60 appropriate to the time t of the star or landmark sighting and its position (AZ, EL) in the instrument aperture 23, 22.
In order to serve as a meaningful quality check of the entire range of the compensation signal 60, the stars and landmarks should be well distributed over the entire instrument 1, 2 field of view.
The OGE provides a continual quality monitoring of the compensation system. With a properly operating system, because of the frequency of star sightings and attitude model updates, the measurement residuals will normally be small. Any continued increase in these residual above a controllable preselected threshold causes an alert message to be genPrated.
The OGE provides an additional continuous quality check on the invention as follows. As part of the data stream continuously telemetered to the ground is sent the compensation signal 60; the AZ, ~L of each mirror 33, 32; and the error (feedback) signal leaving each servo 44, 48, 54, 58. the OGE computer has stored therein a duplicate of the compensation algorithm embedded within processor 38. This duplicate algorithm takes the telemetered information and calculates AZ and ~ EL, then compares these calculated values of ~ AZ and ~ EL
with the telemetered values of ~ AZ and ~ EL. They should coincide if the system is operating properly.
The present invention addresses the problem of long-term errors that impact image registration. Short-term stability errors are also present, and arise from two sources: spacecraft platform stability errors, and ~0 scan repeatability errors. The spacecraft platform stability errors have three sources: motion of the spacecraft in pitch and roll due to Qarth sensor 19 noise (the primary error), errors from mirror 33, 32 motive interaction, and the effects of solar array 11 drive operation.
Scan repeatability errors are due to fixed pattern noise of the inductosyn servos 44, 48, 54, 58; one-cycle errors; sine/cosine unbalances; second harmonic erxors;
bearing noises; bearing friction; wire drag; and servo transient errors.
The mirror 33, 32 interaction errors and solar array ll drive effect errors can be compensated for by the invention known as "Pointing Compensation System for Spacecraft Instruments: described in U.S. Patent No~
4,687,161, commonly assigned with the instant invention.
Spacecraft motion compensation logic 25 of said patent ~2~
application can be implemented as part of processor 33 descriDed herein.
~ he above description is included to illustrate the operation of the 2referred embodiments and is not meant S to limit the scope of the invention. The scope of the invention is to be limited only by the Eollowing claims. From the above discussion, many variations will be apparent to one skilled in the art that would yet be encompassed by the spirit and scope of the 10 invention.
What is claimed is:
Claims (9)
1. Apparatus for limiting the error, to within a preselected limit, in the angular separation of corresponding pixels (with respect to each other) of repeated images of the same selected imaging area of a spacecraft camera, said apparatus comprising:
on board the spacecraft, means for pointing the camera towards a scene external to the spacecraft;
at a location external to the spacecraft, means for determining long-term motion perturbations on the orbit and attitude of the spacecraft; and coupled to said determining means, means for periodically communicating to a computer on board the spacecraft a representation of said long-term motion perturbations;
whereby the computer generates an image registration compensation signal in response to said representation,- and sends said compensation signal to the pointing means.
on board the spacecraft, means for pointing the camera towards a scene external to the spacecraft;
at a location external to the spacecraft, means for determining long-term motion perturbations on the orbit and attitude of the spacecraft; and coupled to said determining means, means for periodically communicating to a computer on board the spacecraft a representation of said long-term motion perturbations;
whereby the computer generates an image registration compensation signal in response to said representation,- and sends said compensation signal to the pointing means.
2. The apparatus of claim 1 wherein the spacecraft further comprises a second camera;
wherein the compensation signal maintains, within a preselected limit, the angular separation of corresponding pixels of repeated images of the same selected imaging area of the second camera.
wherein the compensation signal maintains, within a preselected limit, the angular separation of corresponding pixels of repeated images of the same selected imaging area of the second camera.
3. The apparatus of claim 1 wherein the long-term motion perturbations that are compensated for include at least one of the set of perturbations comprising orbital inclination, orbital eccentricity, yaw error, structural thermal distortion, and earth sensor thermal variation.
4. The apparatus of claim 1 wherein the pointing means comprises:
a mirror disposed to face said external scene;
a gimbal for positioning the mirror in a wanted direction with respect to each of two orthogonal axes; and a drive means for selectively activating the gimbal and thereby partially pivoting the mirror about each of the two axes; wherein the compensation signal is applied to said drive means
a mirror disposed to face said external scene;
a gimbal for positioning the mirror in a wanted direction with respect to each of two orthogonal axes; and a drive means for selectively activating the gimbal and thereby partially pivoting the mirror about each of the two axes; wherein the compensation signal is applied to said drive means
5. The apparatus of claim 4 further comprising apparatus to command the gimbal to scan across the imaging area in a series of scan lines; wherein the representation comprises a set of orbital and attitude coefficients; and the computer further takes into account when calculating the compensation signal real-time information indicating the beginning and end of each scan line, the direction of each scan line, the mirror position, and the camera's operating mode.
6. The apparatus of claim 4 wherein:
the spacecraft telemeters to the external location the compensation signal, the position of the mirror with respect to each axis, and a servo signal by which the drive means activates the gimbal;
the external location, using means identical to those used by the computer, calculates a test compensation signal based on the telemetered mirror position and servo signal; and the external location compares the compensation signal with the test compensation signal as a quality check on the apparatus.
the spacecraft telemeters to the external location the compensation signal, the position of the mirror with respect to each axis, and a servo signal by which the drive means activates the gimbal;
the external location, using means identical to those used by the computer, calculates a test compensation signal based on the telemetered mirror position and servo signal; and the external location compares the compensation signal with the test compensation signal as a quality check on the apparatus.
7. The apparatus of claim 1 wherein the representation comprises a harmonic series having a set of coefficients representing the effects of the sun's apparent rotation about the spacecraft on spacecraft attitude and thermal variations.
8. The apparatus of claim 1 wherein the camera periodically observes stars and landmarks;
the spacecraft telemeters the star and landmark observations to the external location; and the external location incorporates said observations into the representation of the long-term motion perturbations.
the spacecraft telemeters the star and landmark observations to the external location; and the external location incorporates said observations into the representation of the long-term motion perturbations.
9. The apparatus of claim 8 wherein the following method is used to check the quality of the apparatus:
at the external location, generate a model of predicted star and landmark locations;
at the external location, form a set of measurement residuals defined as the difference between predicted locations of the stars and landmarks, on the one hand, and locations of the same stars and landmarks as measured by the camera and telemetered by the spacecraft, on the other hand and compare the measurement residuals with preselected quality thresholds.
at the external location, generate a model of predicted star and landmark locations;
at the external location, form a set of measurement residuals defined as the difference between predicted locations of the stars and landmarks, on the one hand, and locations of the same stars and landmarks as measured by the camera and telemetered by the spacecraft, on the other hand and compare the measurement residuals with preselected quality thresholds.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US06/860,373 US4688091A (en) | 1986-05-06 | 1986-05-06 | Spacecraft camera image registration |
| US860,373 | 1992-03-30 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1266111A true CA1266111A (en) | 1990-02-20 |
Family
ID=25333077
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA000525044A Expired - Lifetime CA1266111A (en) | 1986-05-06 | 1986-12-11 | Spacecraft camera image registration |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US4688091A (en) |
| EP (1) | EP0245562B1 (en) |
| JP (1) | JPH065171B2 (en) |
| CA (1) | CA1266111A (en) |
| DE (1) | DE3684016D1 (en) |
Families Citing this family (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IE61778B1 (en) * | 1989-01-04 | 1994-11-30 | Emyville Enterprises | Image processing |
| US5204818A (en) * | 1990-05-22 | 1993-04-20 | The United States Of America As Represented By The Secretary Of The Air Force | Surveying satellite apparatus |
| JP2535246B2 (en) * | 1990-07-18 | 1996-09-18 | 宇宙開発事業団 | Rendezvous Maneuver Retry / Recovery Methods |
| DE69204345T2 (en) * | 1991-04-16 | 1996-04-25 | Nippon Electric Co | Satellite twin image acquisition for imaging and control. |
| FR2678088B1 (en) * | 1991-06-21 | 1995-03-03 | Thomson Trt Defense | METHOD AND DEVICE FOR CONTINUOUSLY RECALIBRATING IMAGES IN PANORAMIC SLEEP. |
| CA2083203C (en) * | 1991-11-19 | 1996-10-29 | Riichi Nagura | Image data transmission system capable of obtaining a high resolution stereo image with reduced transmission data |
| US5365269A (en) * | 1992-10-22 | 1994-11-15 | Santa Barbara Instrument Group, Inc. | Electronic camera with automatic image tracking and multi-frame registration and accumulation |
| US5430806A (en) * | 1993-09-07 | 1995-07-04 | Loral Vought Systems Corporation | System for changing perspective of 3-D images obtained from reflected energy signals |
| US6285395B1 (en) * | 1993-11-18 | 2001-09-04 | Hughes Electonics Corporation | Earth sensor for satellite |
| RU2153700C2 (en) * | 1995-04-17 | 2000-07-27 | Спейс Системз/Лорал, Инк. | Orientation and image shaping control system (design versions) |
| US6000661A (en) * | 1996-10-16 | 1999-12-14 | Space Systems/Loral, Inc. | Autonomous spacecraft payload base motion estimation and correction |
| US5864131A (en) * | 1996-12-16 | 1999-01-26 | Motorola, Inc. | System and method for accurate geolocation of images |
| US5978716A (en) | 1997-05-28 | 1999-11-02 | Space Systems/Loral, Inc. | Satellite imaging control system for non-repeatable error |
| US6504502B1 (en) | 2000-01-07 | 2003-01-07 | Hughes Electronics Corporation | Method and apparatus for spacecraft antenna beam pointing correction |
| US20050007273A1 (en) * | 2003-07-11 | 2005-01-13 | The Boeing Company | Method and apparatus for prediction and correction of gain and phase errors in a beacon or payload |
| US7274329B2 (en) * | 2003-07-11 | 2007-09-25 | The Boeing Company | Method and apparatus for reducing quantization-induced beam errors by selecting quantized coefficients based on predicted beam quality |
| US7268726B2 (en) * | 2003-07-11 | 2007-09-11 | The Boeing Company | Method and apparatus for correction of quantization-induced beacon beam errors |
| US7260456B2 (en) * | 2004-01-05 | 2007-08-21 | The Boeing Company | Pixel-frequency slews and filters for star data measurements |
| US7310578B2 (en) * | 2004-01-09 | 2007-12-18 | The Boeing Company | Fast access, low memory, pair catalog |
| US7136752B2 (en) * | 2004-01-09 | 2006-11-14 | The Boeing Company | Method and apparatus for on-board autonomous pair catalog generation |
| FR2899344B1 (en) * | 2006-04-03 | 2008-08-15 | Eads Astrium Sas Soc Par Actio | METHOD FOR RESTITUTION OF MOVEMENTS OF THE OPTICAL LINE OF AN OPTICAL INSTRUMENT |
| US9091552B2 (en) | 2011-10-25 | 2015-07-28 | The Boeing Company | Combined location and attitude determination system and methods |
| US10189580B2 (en) | 2017-06-16 | 2019-01-29 | Aerobo | Image stabilization and pointing control mechanization for aircraft imaging systems |
| KR102013647B1 (en) * | 2017-11-24 | 2019-10-21 | 한국항공우주연구원 | Method for satellite image processing and recording medium Therefor |
| US20220329736A1 (en) * | 2021-04-12 | 2022-10-13 | Raytheon Company | Payload yaw rotation for focal plane cross-track columnar scan sampling |
| CN113589318B (en) * | 2021-07-30 | 2023-09-19 | 上海无线电设备研究所 | Simulation method for entrance pupil radiation image of satellite-borne infrared staring camera |
| CN114815128B (en) * | 2022-05-19 | 2023-04-07 | 中国科学院长春光学精密机械与物理研究所 | Micro-nano remote sensing camera on-orbit real-time imaging adjusting system |
Family Cites Families (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3223777A (en) * | 1962-11-26 | 1965-12-14 | Jack A Crawford | Scanner system |
| US4300159A (en) * | 1966-09-30 | 1981-11-10 | Nasa | Scanner |
| US3769710A (en) * | 1969-04-01 | 1973-11-06 | R Reister | Electronic celestial navigation means |
| US3676581A (en) * | 1971-02-01 | 1972-07-11 | Us Navy | Optical scanning spacecraft system |
| US3716669A (en) * | 1971-05-14 | 1973-02-13 | Japan Eng Dev Co | Mapping rectifier for generating polarstereographic maps from satellite scan signals |
| US3859460A (en) * | 1972-11-27 | 1975-01-07 | Baird Atomic Inc | Passive image stabilization system |
| US3952151A (en) * | 1973-08-13 | 1976-04-20 | Trw Inc. | Method and apparatus for stabilized reproduction of remotely-sensed images |
| US4012018A (en) * | 1973-10-04 | 1977-03-15 | The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration | All sky pointing attitude control system |
| JPS6031241B2 (en) * | 1976-07-13 | 1985-07-20 | 三菱電機株式会社 | Satellite attitude determination device |
| US4439788A (en) * | 1982-01-27 | 1984-03-27 | Ball Corporation | Video imaging apparatus having a pliant clock |
| US4602375A (en) * | 1982-06-11 | 1986-07-22 | Communications Satellite Corporation | Onboard clock correction by means of drift prediction |
| GB2149258B (en) * | 1983-11-04 | 1987-03-11 | Ferranti Plc | Image correction system |
| US4593317A (en) * | 1984-08-13 | 1986-06-03 | The United States Of America As Represented By The Secretary Of The Navy | Moving scene display for passive radiation-wave imaging system |
| US4639774A (en) * | 1985-06-21 | 1987-01-27 | D. L. Fried Associates, Inc. | Moving target indication system |
-
1986
- 1986-05-06 US US06/860,373 patent/US4688091A/en not_active Expired - Lifetime
- 1986-12-11 CA CA000525044A patent/CA1266111A/en not_active Expired - Lifetime
- 1986-12-15 EP EP86309780A patent/EP0245562B1/en not_active Expired - Lifetime
- 1986-12-15 DE DE8686309780T patent/DE3684016D1/en not_active Expired - Lifetime
- 1986-12-29 JP JP61315938A patent/JPH065171B2/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| DE3684016D1 (en) | 1992-04-02 |
| EP0245562B1 (en) | 1992-02-26 |
| JPH065171B2 (en) | 1994-01-19 |
| JPS62263407A (en) | 1987-11-16 |
| EP0245562A3 (en) | 1989-02-08 |
| EP0245562A2 (en) | 1987-11-19 |
| US4688091A (en) | 1987-08-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA1266111A (en) | Spacecraft camera image registration | |
| JP2598820Y2 (en) | Image navigation support system for satellite with camera | |
| CA1297563C (en) | Satellite camera image navigation | |
| US5412574A (en) | Method of attitude determination using earth and star sensors | |
| US8301377B2 (en) | Image navigation method using parametric systematic error correction | |
| US6023291A (en) | Satellite camera attitude determination and image navigation by means of earth edge and landmark measurement | |
| US5899945A (en) | Attitude control and navigation system for high resolution imaging | |
| US5107434A (en) | Three-axis spacecraft attitude control using polar star sensor | |
| EP0974516A2 (en) | Spacecraft camera image navigation and registration | |
| US20080063270A1 (en) | Method and Apparatus for Determining a Location Associated With an Image | |
| US5852792A (en) | Spacecraft boresight calibration filter | |
| JPH08190161A (en) | Focal-plane-array imaging device for satellite | |
| US5978716A (en) | Satellite imaging control system for non-repeatable error | |
| Kamel | GOES image navigation and registration system | |
| Markley et al. | Attitude control system conceptual design for geostationary operational environmental satellite spacecraft series | |
| US5821526A (en) | Star scanning method for determining the line of sight of an electro-optical instrument | |
| KAMEL et al. | GOES IM image motion compensation system | |
| Kamel et al. | Star sightings by satellite for image navigation | |
| MARKLEY et al. | Attitude control system conceptual design for the GOES-N spacecraft series | |
| KR100314857B1 (en) | The remote control investigation system for aircraft | |
| Marc et al. | ESA remote-sensing satellites: A cross-correlation of in-flight attitude pointing performances | |
| Kamel et al. | Satellite camera image navigation | |
| Weaver et al. | ERBE and CERES broadband scanning radiometers | |
| Yang et al. | Study on automatic airborne image positioning model and its application in FY-3A airborne experiment | |
| KR20080033287A (en) | Method and apparatus for determining a location associated with an image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MKLA | Lapsed |